This
Message Is From an External Sender
|
This
message came from outside your
organization.
|
|
|
OK. We still have the other PR as well
though.
Sure, that
approach works. We’ll go ahead with this latest staged TCK
then.
The issue you
mention below sounds like what Petr just put in an update to
the TCK to switch away from, and is why I did the re-spin, so
hopefully that will run cleanly for you now.
It may be we have to
do test challenges post ballot using the TCK challenge
process. We haven’t got to the bottom of it yet but some
of the tests which pass a lookedup ManagedThreadFactory
out of the originating component and then expect to
This
Message Is From an External Sender
|
This
message came from outside your
organization.
|
|
|
It may be we have
to do test challenges post ballot using the TCK challenge
process.
We haven’t got to
the bottom of it yet but some of the tests which pass a
lookedup ManagedThreadFactory out of the originating component
and then expect to use the originating component namespace
seem to be contrary to the behaviour of the original GlassFish
Concurrency RI.
Steve
Petr, if there
are other TCK issues that you were going to open, it would be
good to get them open and looked at right. We came very close
to requesting spec approval this morning and would have done
so if it weren’t for the Open Liberty publish of the results
getting hung up. Even after this latest one, we’ve already
started to prepare the next attempt at running and publishing,
and we shouldn’t be doing that if you will be opening other
issues. Also, we need to consider what this will mean for EE
10. My understanding is that if we don’t request spec
approval in the next few days, then we have missed EE 10.
I’ll take a look at
the job to see what it is doing. I now Petr has some other
TCK challenges he is working on. From: Nathan Rauh <nathan.rauh@xxxxxxxxxx> Sent: 26 April
2022 18:03 To: cu developer discussions <cu-dev@xxxxxxxxxxx>;
This
Message Is From an External Sender
|
This
message came from outside your
organization.
|
|
|
I’ll take a look
at the job to see what it is doing. I now Petr has some other
TCK challenges he is working on.
Petr found a
copy/paste error that needed to be fixed in the TCK, and so I
had to re-build the TCK.
Both builds claimed to be successful, however, it appears that
the Promote Staged TCK didn’t actually overwrite the artifacts
at the promoted location
Staged:
https://download.eclipse.org/ee4j/cu/jakartaee10/staged/eftl/concurrency-tck-3.0.0.info
2fe905b8adfab903a6c5954453e67fbc3ed583b0e313edef7f09d18b7eddd0ee
concurrency-tck-3.0.0-dist.zip
-rw-r--r--.
1 jenkins 1001550000 398136 Apr 26 16:27
concurrency-tck-3.0.0-dist.zip
Promoted:
https://download.eclipse.org/ee4j/cu/jakartaee10/promoted/eftl/concurrency-tck-3.0.0.info
93fdfc808fff4e2fbd52883c7378b369d8ee81c87964ca8d3b6dfc20eb03e1d4 concurrency-tck-3.0.0-dist.zip
-rw-r--r--. 1 jenkins 1001550000 398134 Apr 25 18:21 concurrency-tck-3.0.0-dist.zip
Note the date and
SHA under promoted still match the original, not the updated
from staged.
We can probably
go ahead with the release approval based only on the staged
location, assuming that it will be possible to fix the Promote
Staged TCK build to do an overwrite (Arjan and Steve –
your help is needed with this). However, if we do that, it
will be important that no one submits any further Concurrency
TCK Master Builds or merges changes to the spec (which will
trigger a Concurrency TCK Master Build) in the meantime
because doing so would overwrite the staged copy that we are
attempting to use for approval.
Hi All, Thanks for the
assistance from various people. We now have both staged
and promoted TCK zip files on Eclipse Downloads Project
download area | The Eclipse Foundation Project download
area | The Eclipse Foundation
This
Message Is From an External Sender
|
This
message came from outside your
organization.
|
|
|
Hi All,
Thanks for the
assistance from various people.
We now have both
staged and promoted TCK zip files on Eclipse Downloads
Project download area | The Eclipse
Foundation
Project download area | The Eclipse
Foundation
I have modified
the TCK Build job
Jakarta Concurrency - Concurrency TCK Master Build [Jenkins]
(eclipse.org) to automatically upload to the staged
directory on downloads if the build succeeds. This job is
triggered by a commit in the repo which means that if there is
a change a new TCK zip will appear in the staged directory
with likely a new SHA.
I have also
created a Jenkins job
Jakarta Concurrency - Promote Staged TCK [Jenkins]
(eclipse.org) which copies the zip file from the staged
directory to the promoted directory. This job is manually
triggered so we have control if we want to “promote” a
specific “staged” zip file.
I’m sure these
jobs could be done better if I had a better view of what an
ideal workflow would be for releasing the TCK but should
suffice to get us to ballot.
Any problems give
me a shout.
Steve Millidge